"Wart" on an ASUS power supply

[ebastler]

The power supply of my ASUS laptop has developed a strange "wart" or blister on one side -- see the photo here:



It is a small bump, nearly circular, about 3mm in diameter. It seems to have melted around the perimeter, while the central area still seems to have the original texture of the plastic. (But that might be a misinterpretation; maybe it has just developed a similar speckle pattern when the molten plastic hardened again.)

I don't know when and how this has happened. Thermal overheating from the inside seems unlikely for such a sharply localized defect? But I can't recall any recent circumstances where external heat or solvents could have caused this. I don't want to crack the unit open, since it continues to work, but am concerned that it may be on the brink of failing.

Has anybody seen this before? The power supply is 3 years old and came with a UX303L laptop.
Model APD-650W C
Input 100-240V~ 50-60Hz  1.5A
Output 19V 3.42A

[Haenk]

This PSU is about to fail and is certainly not safe for use any more. Best case is tripping a fuse, worst case might be a small fire.

Bonus points for cracking it open without cutting oneself. Most likely a dried out electrolytic condenser causing havoc (so it is fixable if still working, but impossible to get it back into the casing, so better just get a new one.)

[wraper]

Have some doubts if it's really melted plastic. Looks somewhat unnatural to form small protruding spot, without any plastic deformation around it. Not to say there should be metal shield around PCB. Could it be just a drop of some liquid that solidified? I suggest running this PSU and checking if that spot actually gets that hot and check if plastic becomes soft. Also try washing it off just in case.

[james_s]

Yeah I don't think that's melted from inside. Either something has dripped on it or it is a molding defect in the plastic. Are you sure it wasn't always there?

[ebastler]

Thank you all for the responses so far. So at the moment the vote is divided 1:2 between "it's a death trap" and "it's some cosmetic blip on the outside". 

I am pretty sure the spot has not been there all the time, but is a recent addition. Under a strong magnifier, it looks very much like molten plastic, and I can't scrape it off. (I don't have a microscope with camera, but will try to get a photo through the magnifier later.)

When I plug the power supply in, the case does not become noticeably hot at that spot. I did use the power supply on trains a couple of times recently, which might (or might not) provide "dirty" power on their outlets? But that sounds a bit far-fetched to me. And, and as stated earlier, I struggle to see how the heating and melting would stay so localized if it were to originate from some internal heat source.

Still open for further votes! 

[Gyro]

Another vote for 'You never noticed it before or something has been dripped on it externally'. If there is no localised heating at that point during operation then I don't see how it could have formed from an internal cause.

[magic]

Well, one thing I have seen in a (shitty) PSU is a drop of molten solder falling off the PCB, but a different thing I have never seen is such a drop of solder causing any visible signs externally.

Does anything rattle inside?

[Twoflower]

Hard to see, but could that caused from the outside? Something hot or pushing strong enough (e.g. a rivet) imprinted itself on the housing?

Causing such exact and sharp forms to form through few mm of plastic seems to me very unlikely. Unless it is already very close to the surface. But I still would expect some additional signs of something pushing through the case.

[wraper]

I am pretty sure the spot has not been there all the time, but is a recent addition. Under a strong magnifier, it looks very much like molten plastic, and I can't scrape it off.

What if you use wet wipe? Or some solvent.

[SMdude]

Or give it a scrape with a small screwdriver.
It does look like some glue or something like that.

[ebastler]

Here's an image taken through a magnifier. Still not as good as if you can look at it hands-on, varying the direction of view and lighting, but I hope it gives a better idea.

The whole structure, including the glossy outer ring, is bulging outward slightly. So it appears that there is more material in total than there was in the original flat surface. But it certainly does not look like some foreign, added substance; the blob seems autogeneous. Mild solvent (isopropanol) doesn't have any effect, and I can't scrape anything off.

Maybe a drop of more aggressive solvent has hit the surface recently -- but could it have caused swelling? Looking at the glossy outer ring, I had also thought about someting like a hot rivet having touched the surface, but again, why would that make the whole structure bulge outward? Also, the ring is not exactly circular.

Anyway, with the majority opinion swinging stronlgy towards "external impact", I am inclined to just keep using the PSU. I certainly don't mind the wart, as long as I don't have to be concerned about electrocuting my computer or myself, or setting things on fire...

P.S.: Nothing rattles inside either!

[Twoflower]

The magnified picture shows a regular pattern in in the material just inside of the ring. So I really thing that came from an outside event. Something imprinted its pattern from the outside.

[schmitt trigger]

With a little imagination, it could be used as a prop in a low budget horror movie.

[ebastler]

With a little imagination, it could be used as a prop in a low budget horror movie.

That thought crossed my mind... "Nah, it's not gonna catch fire. But some critter will break through, poke its head out and eat me."

[Electro Detective]

If it tests and regulates fine under load, no hot spots, no buzzing, crackling or hissing,
it's probably ok, but keep it off carpets and beds 

OTOH if it was my wart, I'd carefully crack it open and take a looksee anyway
there's probably a Youtube or post somewhere showing how to do that particular psu without trashing it too much 

If it's ok or fixable and or 'improvable', then after reassembly there's no more guesswork or crossed fingers about it 


Yes this is a clever ruse    to get you to crack it open so we can have a tech perv 

[Haenk]

Just as a quick note: I have sold 10s of thousands external drives using these types of encapsuled PSUs - and have seen a lot (well, maybe 20) of cases with those "bubbles" appearing, all from failed PSUs, even with burnt in holes (which shouldn't happen at all, but still *does* happen).
If the electrolyt dries up, the cap can get really hot, way beyond the melting point of the plastic material.
So I certainly do suggest to replace it immediatly.

[magic]

I presume the output caps would be the first to fail, if any, so this should be detectable with a scope.

Which is to say, is this the Tom's Hardware forum that we are casting votes instead of just testing the damn thing?

[Electro Detective]

Break it open to see the result of cheap no name and or fake parts, trapped in a sealed nice looking  plastic box
that can't breathe or expel heat

Yet another successful Marketing and Bean Counter Department 'Electronical Engineering' collaboration 

[ebastler]

Alright, guys. I am back home now, and will scope the output voltage under load tonight, and will keep tabs of the case temperature while doing so.

Unless these measurements show that something is wrong, I am not inclined to crack the PSU's case open. The thing is ultrasonically welded, with a complex shape of the weld (case bottom and lid, plus the EU plug insert). Even if I manage to mill it open and glue it back together without totally ruining it, that is bound to make the whole thing less safe than it is now...

Haenk -- your feedback that you have seen similar damage before is what I was looking for. Does that include similar-looking, sharply localized "blisters"? I still struggle to imagine how heat spreads through the plastic without diffusing laterally and creating a wider bump or valley. 

[fzabkar]

Can you drill an inspection hole with a sheet metal drill? It shouldn't be too difficult to patch the hole without compromising safety.

[james_s]

Got a magnifying desk lamp that might have caught the sun? Kids with a magnifying glass or a powerful laser pointer? It looks a bit like what happens if you focus the sun into a small spot on plastic.

[magic]

If kids got their hands on a laser capable of doing that, there would probably be more signs of the event all over the place

[ebastler]

Can you drill an inspection hole with a sheet metal drill? It shouldn't be too difficult to patch the hole without compromising safety.

Hmm, but what if there is sheet metal for shielding right behind the plastic? I would hate to spread metal shavings inside the PSU. Don't think I am ready to take that risk.

I did monitor the PSU's temperature while it fully charged an empty battery. It gets nice and warm, as it always has, but nothing unusual or concerning. And the area of the suspicious spot does not get hot; if anything, it is cooler than other areas on the side of the PSU.

I still need to rig up an adapter to allow me to measure the 19V output voltage while charging, or cobble a suitable dummy load together. I will do that soon, and in the meantime won't use the PSU unattended.

[Twoflower]

I still fail to understand how the regular structure (marked in the attachment) can happen if something from the inside melts the housing. The structure seems to run around the ring and is extremely regular and does not match the normal structure of the housing.

[ebastler]

I still fail to understand how the regular structure (marked in the attachment) can happen if something from the inside melts the housing. The structure seems to run around the ring and is extremely regular and does not match the normal structure of the housing.

Yes, it's strange. The outer ring is completely smooth (but bulging outward slightly). Then there is an inner ring with quite regular radial structures -- wrinkles from the plastic melting and then hardening/contracting? It is not as regular as an embossed structure would probably be; and as mentioned earlier, the whole structure is not precisely round, but slightly oval.

A crop circle appeared on my power supply!